WiFi RSSI preprocessing library for Android

Fingerprinting is one of the most popular indoor positioning approach, which enables the usage of heuristic and data mining methods for position estimation. Many fingerprinting-based indoor positioning algorithms use the WiFi Received Signal Strength Indication data and Horus System showed that client side preprocessing could improve the stability, and accuracy of the positioning system. Although filtering techniques are well-studied and widely applied in data sciences, their applications in the field of indoor positioning is limited. Providing a general and flexible filtering library would facilitate the development of client applications for indoor positioning systems. This paper presents a Java library that allows the application of filtering methods on Android platform. Flexibility and extendability were concerned during the design of the library, which currently contains three different filtering methods, namely Horus, Static Time Window and Dynamic Time Window filters. An Android applications was developed to demonstrate the applicability of the filtering library in real life scenario.